TimeMarker: benchmark results
Provider: Other.
Aggregate Rating 1520 ± 42, rank #2063 of 10916 rated models, from 23 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MLVU-Test - Topic Reasoning | 85.7 | Multiple-choice accuracy (%) | 89.7 |
| MLVU-Test | 49.2 | Multiple-choice accuracy (%) | 85.3 |
| MLVU-Test - Needle QA | 65 | Multiple-choice accuracy (%) | 85.3 |
| MLVU-Test - Action Count | 26.7 | Multiple-choice accuracy (%) | 82.4 |
| MLVU-Test - Anomaly Recognition | 53.9 | Multiple-choice accuracy (%) | 79.4 |
| MLVU-Test - Tutorial QA | 37.2 | Multiple-choice accuracy (%) | 79.4 |
| MLVU-Test - Ego Reasoning | 49.1 | Multiple-choice accuracy (%) | 76.5 |
| MLVU-Dev | 63.9 | Multiple-choice accuracy (%) | 73.7 |
| LongVideoBench | 59.6 | Test Total Accuracy (%) | 70.6 |
| MLVU-Test - Action Order | 31.4 | Multiple-choice accuracy (%) | 70.6 |
| MLVU-Test - Sports QA | 41.7 | Multiple-choice accuracy (%) | 70.6 |
| MLVU-Test - Plot QA | 52 | Multiple-choice accuracy (%) | 69.1 |
Interactive version: theaggregate.ai/model?slug=timemarker · How It Works · Data refreshed daily, snapshot 2026-11-08.